See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/373293857

Smart Food Monitoring System Based on IoT and Machine Learning

Conference Paper · June 2023

DOI: 10.1109/NCIM59001.2023.10212608

CITATIONS READS
15 719

3 authors:

Md. Johirul Islam Tutul Mahanur Alam

Bangladesh University of Business and Technology Bangladesh University of Business and Technology
2 PUBLICATIONS 15 CITATIONS 2 PUBLICATIONS 15 CITATIONS

SEE PROFILE SEE PROFILE

Md. Anwar Hussen Wadud

Sunamgonj Science and Technology University
57 PUBLICATIONS 413 CITATIONS

SEE PROFILE

All content following this page was uploaded by Md. Anwar Hussen Wadud on 22 April 2024.

The user has requested enhancement of the downloaded file.

 2023 International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM)
16-17 June 2023, Gazipur-1707, Bangladesh

Smart Food Monitoring System Based on IoT and

Machine Learning
Md. Johirul Islam Tutul ∗ , Mahanur Alam† , Md. Anwar Hussen Wadud‡
Department of Computer Science and Engineering
Bangladesh University of Business and Technology∗†‡ , Mawlana Bhashani Science and Technology University‡
[email protected]∗ , [email protected]† , [email protected]‡

Abstract—The monitoring of food quality is crucial in ensuring making it accessible for both commercial and personal use [2].
food safety and reducing food waste. In this paper, we present the The system includes a web-based dashboard that displays real-
design and implementation of a smart food monitoring system time data and provides alerts when food items reach a certain
that utilizes Internet of Things (IoT) and Machine Learning
(ML) technologies. The system includes IoT devices such as threshold level. The dashboard can be accessed remotely using
temperature and humidity sensors, as well as gas concentration a mobile app, making it convenient for users to monitor their
sensors, to continuously monitor food items. The data collected food from anywhere. The proposed system also includes a
by these sensors is analyzed using machine learning algorithms machine learning algorithm that analyses the collected data
to determine the freshness of the food. The system also features to predict the freshness of the food. The algorithm is trained
a web-based dashboard that displays real-time data and alerts
when the food items reach a certain threshold level. The dash- using historical data, and its performance is evaluated using
board can be accessed remotely using a mobile app, making it various food items. The results show that the algorithm is suc-
easy for users to monitor their food from anywhere. The system’s cessful in predicting the freshness of food with high accuracy.
effectiveness was evaluated using various food items, and the In summary, the proposed IoT-based food quality monitoring
results showed that it successfully predicted the freshness of system is a promising solution for the food industry, providing
food with high accuracy. The proposed system has significant
potential applications in the food industry, reducing food waste real-time monitoring of food items and accurate predictions
and ensuring that consumers can access fresh and healthy food. of food freshness. The system’s affordability, ease of use, and
Additionally, the system provides an efficient way of tracking food accessibility make it ideal for both commercial and personal
inventory, reducing costs and increasing efficiency. The system’s use, reducing food waste, improving food safety, and ensuring
real-time monitoring and alerts are ideal for personal use in access to fresh and healthy food.
homes and kitchens. In conclusion, the proposed smart food
monitoring system using IoT and machine learning provides a Overall, our food monitoring system provides a compre-
reliable and cost-effective solution for monitoring the freshness hensive solution to the issue of food safety and waste. By
of food. The system has significant potential applications in the
food industry and personal use, reducing food waste, increasing combining real-time data collection with easy installation and
efficiency, and ensuring access to fresh and healthy food. accurate predictions, our system provides users with a pow-
erful tool for monitoring the quality of their food items. The
Keywords—Food Safety, Food Quality, Temperature & Hu-
midity Monitoring, Gas Concentration Monitoring, Internet of combination of machine learning algorithms with IoT enables
Things, Machine Learning. the food monitoring system to provide accurate predictions of
food freshness in real time. By collecting and analyzing data
I. I NTRODUCTION from multiple sensors, the system can detect changes in the
quality of food items and alert the user before they spoil. This
The monitoring of food quality has been a long-standing helps to reduce food waste and improve food safety, making
challenge for the food industry. In the past, food quality it a valuable tool for both commercial and personal use [3],
monitoring systems have been limited to manual inspections, [4].
which can be time-consuming and prone to errors. With the
emergence of IoT and machine learning technologies, food The scenario in Fig. 1 illustrates using a methane sensor
quality monitoring has become more efficient and accurate [1]. to detect the spoilage of two apples, and then sending a real-
These technologies enable real-time monitoring of food items, time alert to a user’s mobile device using Blynk to indicate
as well as data analysis to predict the freshness of food. The the potential danger. This system can be described as follows:
performance of machine learning algorithms in food quality First, a methane sensor is placed near the two apples to
monitoring has been well-documented. These algorithms can detect any methane gas emissions resulting from the apples
analyze data from sensors such as temperature, humidity, and spoiling. Methane gas is a by-product of the decomposition
gas concentration, to predict the freshness of food items. process and can be used as an indicator of spoilage. The
Various studies have shown that machine learning algorithms methane sensor is connected to a microcontroller, which is
can accurately predict the freshness of food items, making programmed to read the sensor data and analyze it for signs
them a promising solution for food quality monitoring. of spoilage. If the methane levels detected by the sensor
exceed a certain threshold, this indicates that the apples are
In this paper, we propose an IoT-based food quality mon- spoiling and emitting methane gas. The microcontroller is then
itoring system that combines temperature, humidity, and gas connected to a wireless module, such as a Wi-Fi or Bluetooth
concentration sensors to provide real-time monitoring of food module, which allows it to communicate with the user’s
items. The system is designed to be affordable and easy to use, mobile device. The mobile device must have the Blynk app
979-8-3503-1600-1/23/$31.00 ©2023 IEEE
 II. M OTIVATIONAL BACKGROUND AND R ELATED W ORKS
A. Motivation
The motivation behind developing a food monitoring system
is to ensure food safety and reduce food waste. Every year,
millions of people around the world suffer from foodborne
illnesses caused by consuming contaminated food. In addition,
a significant amount of food is wasted due to spoilage, leading
to economic losses and negative environmental impacts [8]. By
implementing a food monitoring system, it becomes possible
to detect food quality issues early on and prevent the distribu-
tion of contaminated food. This can save lives, reduce health-
care costs, and protect the reputation of food manufacturers
and retailers [9]. Additionally, the food monitoring system
can help reduce food waste by providing accurate information
on the freshness of food items. By knowing exactly when a
food item is going bad, users can consume it before it spoils,
reducing waste and saving money. Overall, the motivation
behind developing a food monitoring system is to promote
food safety and sustainability by detecting food quality issues
Fig. 1. Mechine learning and IOT combinedly detect the spoilage food
early on and reducing food waste [10].

B. Related Works
installed, which allows the user to receive real-time alerts and In the research by Sundaravadive et al. [11] proposed a
notifications based on sensor data. When the microcontroller smart food monitoring system that uses deep learning and IoT
detects methane levels have exceeded the threshold, it sends a for food safety. The system collects data on the temperature,
signal to the mobile device via the wireless module [5]. This humidity, and weight of food items and uses a deep learning
signal triggers an alert in the Blynk app, which displays a algorithm to predict the probability of food spoilage. The
message to the user indicating that the apples are spoiling and system consists of various sensors and IoT devices that collect
that they should be discarded immediately. In this way, the data in real time. The system required a large amount of
system allows the user to be alerted to the danger of spoiling labelled data for training the deep learning model. The system
apples in real-time, allowing them to take action to avoid any is also limited to predicting food spoilage based on a few
potential health risks [6], [7]. The system can be extended environmental factors and does not consider other factors
to other perishable foods, such as meat or dairy products, by such as microbial growth [12]. The paper [5] presented a
using appropriate sensors to detect the specific gases produced smart food inventory system that uses IoT and blockchain
during spoilage. The main contributions of this paper are: technology for supply chain management. The system tracks
the movement of food items from farm to table, ensuring food
• Improved food safety: The system can continuously safety and reducing food waste. The system consists of various
monitor food quality parameters such as temperature, IoT devices and Blockchain [13]–[15] technology that tracks
humidity, and other environmental conditions during food the movement of food items from farm to table. The system
storage and transportation, which can reduce the risk of uses smart contracts to ensure the quality and safety of food
spoilage and contamination, leading to improved food items. The system requires a high level of technical expertise
safety. and resources to implement. The blockchain technology used
• Real-time monitoring: The system can provide real-time in the system can also be limited by its scalability, security,
monitoring of food conditions and alert users in case of and regulatory challenges. Wireless sensor network-based food
any deviations from the optimal conditions. This can help quality monitoring system that uses machine learning [16],
prevent food spoilage and ensure food freshness. [17]. The system collects data on food quality indicators such
as pH, temperature, and humidity and uses machine learning
• Efficient inventory management: The system can provide algorithms to predict the quality of food items. It consists of
real-time data on food inventory, allowing for better a wireless sensor network that collects data on food quality
management of food stocks, reducing food waste, and indicators in real-time. The data is prepossessed and fed into
optimizing food supply chain management. a machine-learning model for prediction [18]. It is limited by
the accuracy of the machine learning algorithms used and
Organization: The paper is organized as follows: section the availability of labeled data for training the model [6].
II describes the motivational background and existing research IoT-enabled smart food storage system for food safety and
work on IoT-based smart food monitoring systems. We also shelf life prediction. The system collects data on temperature,
provide an overview of machine learning techniques that can humidity, and gas concentrations in food storage areas and
be used to analyze food consumption patterns and optimize uses machine-learning algorithms to predict the shelf life of
inventory management in section III. Section IV describes food items. The system consists of various sensors and IoT
the experimental analysis and section V concludes the paper. devices that collect data in real time. The data is preprocessed
and fed into a machine-learning model for prediction [19]. III. M ETHODOLOGY
The system is limited by the accuracy of the machine learning Our smart food monitoring system with IoT collects data
algorithms used and the availability of labeled data for training from various sensors and devices such as temperature sensors,
the models. The system is also limited to predicting shelf life humidity sensors, weight sensors, and cameras. This data is
based on a few environmental factors and does not consider sent to a central server, where it is processed using the data
other factors such as microbial growth [20]. processing algorithm. First, the raw data is loaded into the
On the another study, the paper [21] focused a smart food system, which may be in different formats and structures
safety management system that uses IoT and cloud computing. depending on the sensor or device. The data is then pre-
The system collects data on food safety indicators such as processed to handle any missing values and outliers and to
temperature, humidity, and gas concentrations and uses cloud normalize the data. For example, if there is missing data from
computing to analyze the data and provide real-time alerts for a sensor, the data processing algorithm can impute the missing
food safety violations. The system consists of various sensors data or remove the corresponding samples entirely. Next, the
and IoT devices that collect data in real-time. The data is data is analyzed to gain insights and understand patterns in
transmitted to the cloud for analysis and real-time alerts. The the data. For example, the data processing algorithm can
system is limited by the availability of reliable internet connec- compute summary statistics such as the mean and variance
tivity and the need for cloud computing resources. The system of temperature and humidity readings or create visualizations
is also limited to predicting food safety violations based on a such as scatter plots and histograms. These insights can help
few environmental factors and does not consider other factors identify trends and patterns in the data, such as the times of
such as microbial growth [22]. The paper [23] IoT-based smart day or days of the week when certain foods are consumed
food traceability system for food safety and quality assurance. more frequently. After the data has been analyzed, it can be
The system uses RFID technology to track the movement of transformed to create new features or reduce the number of
food items and provides real-time information on food safety features. For example, the data processing algorithm can create
and quality. The system consists of RFID tags attached to new features, such as the duration of time that certain foods
food items and various IoT devices that collect data on the are stored or the number of times a particular food item has
movement of food items in real time. The system is limited been consumed. The algorithm can also reduce the number of
by the availability of reliable internet connectivity and the need features using techniques such as PCA or t-SNE. Finally, the
for RFID technology. The system is also limited to tracking the processed data is returned to the IoT devices for use in real-
movement of food items and does not consider other factors time monitoring and decision-making [32], [33]. For example,
such as environmental conditions and microbial growth [24]. the data can be used to trigger alerts when certain thresholds
are exceeded, such as when the temperature of a refrigerated
The paper [25] presented a smart food safety management food item rises above a safe level. The processed data can
system that uses IoT and cloud computing. The system collects also be used for predictive maintenance, such as forecasting
data on food safety indicators such as temperature, humidity, when equipment will require maintenance or replacement. The
and gas concentrations and uses cloud computing to analyze predictive models can then be used for real-time monitoring
the data and provide real-time alerts for food safety violations. and decision-making. In our system, if the temperature of a
The system consists of various sensors and IoT devices that refrigerated food item rises above a safe level, the machine-
collect data in real time. The data is transmitted to the cloud learning algorithm can trigger an alert, which can then be
for analysis and real-time alerts. The system is limited by sent to the relevant stakeholders through a mobile app or
the availability of reliable internet connectivity and the need another notification system. In this way, machine learning
for cloud computing resources. The system is also limited to and IoT technologies are combined to enable our smart food
predicting food safety violations based on a few environmental monitoring system to collect, analyze, and act on data in real-
factors and does not consider other factors such as microbial time, improving food safety and quality, reducing waste, and
growth [26]. This research [27], [28] focused on developing improving efficiency [27], [34].
an IoT-based smart framework for evaluating food quality in
restaurants and food outlets, utilizing Fog-Cloud computing 1) Solution: In addition to the motivation for a food mon-
and IoT technology within the context of Industry 4.0. The itoring system, several solutions can be provided through our
framework incorporates a 2-player game-theoretic model in- project. Real-time Dashboard: Our food monitoring system
volving monitoring officials and food managers to optimize provides a real-time dashboard that displays temperature,
decision-making processes. Through comparative simulations, humidity, and gas concentration data for each food item being
the proposed model demonstrates superiority over existing monitored. This dashboard provides users with up-to-date
methodologies in terms of data assessment efficacy, statistical information on the quality of their food items, allowing them
classification analysis, decision-making efficiency, reliability, to make informed decisions about consumption and preventing
and stability. This research contributed to the advancement of the distribution of contaminated food.
food quality evaluation by leveraging IoT and Industry 4.0
• Training: Uses historical data to train a machine learning
concepts, providing an innovative approach to enhance food
algorithm to predict food freshness.
quality management in the industry [29], [30]. The limitation
of this paper is that it does not provide a detailed analysis • Easy Method: Our food monitoring system is designed
of the computational and networking requirements of the to be easy to use and accessible to both commercial
proposed framework, which may be a challenge for real-world and personal users. The system consists of an IoT-based
deployment [31]. device that is easy to install and connect to the Internet.
 Algorithm 1: Proposed Algorithm
Input : raw data D
Output: Processed data D′
1 while Load the raw data D do
2 Preprocess the data:
3 if Convert the data into a suitable format for analysis then
4 Handle missing values (e.g., imputation, removal)
5 Handle outliers (e.g., removal, clipping)
6 Normalize the data (e.g., standardization, scaling)
7 if Analyze the data: then
8 Explore the data using summary statistics and visualizations (e.g., mean, median, variance, histogram,
scatter plot)
9 Identify patterns and relationships in the data
10 Determine which features are relevant for the analysis
11 end
12 else if Transform the data then
13 Feature engineering: create new features from existing features (e.g., adding, multiplying, taking logarithms)
14 Feature selection: choose the most important features for the analysis
15 end
16 else if Return the processed data D’ then
17 end
18 else
19 end

The device is low-cost, making it an affordable solution can help improve food safety and reduce food waste. This
for users of all sizes. system consists of sensors that are installed in the food
storage areas and kitchen appliances to collect data about
• Algorithm: Our food monitoring system uses a machine
food usage, temperature, humidity, and other factors that
learning algorithm as shown in Algorithm 1 to predict
can impact food quality. Machine learning algorithms are
the freshness of food items based on historical data. This
used to analyze this data and provide actionable insights
algorithm combines data from temperature, humidity, and
to users. In this paper, we will discuss the experimental
gas concentration sensors to provide accurate predictions
analysis of the smart food monitoring system based on IoT and
over time. By predicting the freshness of food items, our
machine learning [38]. To conduct an experimental analysis
system can reduce food waste and improve food safety.
for the smart food monitoring system, we need to set up
a test environment that simulates real-world scenarios. This
A. System Architecture test environment should include sensors, IoT devices, and a
The algorithm is a critical component of the food monitoring machine-learning model. The following steps can be followed
system, as it is responsible for predicting the freshness of the to set up the test environment: Perishable foods are prone to
food items. By combining data from temperature, humidity, spoilage and degradation, which can pose significant health
and gas concentration sensors, the algorithm provides accurate risks and financial losses. Traditional methods of monitoring
predictions over time as it is trained with more data [17], [35]. food quality and freshness rely on the human inspection
This improves food safety by reducing the risk of foodborne and manual testing, which are time-consuming and error-
illness, as the system alerts users when food items are reaching prone. To overcome these challenges, we propose a smart
a certain threshold level. Additionally, the algorithm can be food monitoring system that leverages the power of deep
adapted to different food items and environmental conditions, learning to automate the process of identifying food items and
making it a versatile solution for different use cases. Further- predicting their quality and freshness. In this paper, we present
more, the algorithm has the potential to reduce food waste a deep learning algorithm that is trained on a dataset of sensor
by accurately predicting the freshness of food items [36], readings from different types of fruits and vegetables, along
[37]. This is especially important in food manufacturing and with their corresponding spoilage levels. Sensor Installation:
retail, where large amounts of food waste can occur due to Install sensors in the food storage areas and kitchen appliances
uncertainty about freshness issues. The algorithm helps to to collect data about food usage, temperature, humidity, and
reduce this waste by providing real-time monitoring of food other factors that can impact food quality. These sensors can
items, which can help to ensure that they are used or sold be wireless and battery-powered. IoT Device Setup: Connect
before they spoil. the sensors to IoT devices such as Esp8266, which act as the
gateway for data transmission. These IoT devices should be
IV. E XPERIMENTAL A NALYSIS programmed to collect sensor data and send it to the cloud.
To consider, the smart food monitoring system based on Cloud Storage: Set up a cloud storage system to store the
IoT and machine learning is a promising technology that
sensor data. This cloud storage should be scalable and secure.
we use blynk cloud here for monitoring our system data.
The spectra of all food samples in the wave number range
of 12,000–4000 cm-1. Original spectra (log 1/R) as shown
in Fig. 2; spectra after pre-treatment by multiplicative signal
correction (MSC) as shown in Fig. 3. The smart food monitor-
ing system combines IoT, machine learning, and spectroscopy
to ensure food quality and safety. It consists of a methane
sensor, microcontroller, buzzer, and real-time monitoring sys-
tems for Android and PC. Spectroscopy is used to obtain the
spectra of food samples, measuring light intensity at different
wavelengths. The original spectra are represented as log 1/R,
where R is the ratio of reflected/transmitted light intensity
to the incident light intensity. Multiplicative signal correction
(MSC) is applied to enhance spectra quality by reducing
unwanted variations. MSC adjusts spectra mathematically, di-
viding each spectrum by its estimated baseline spectrum. This
normalization improves comparability and reduces influences
like baseline shifts, scattering, particle size, and instrumental Fig. 3. spectra after pretreatment by multiplicative signal correction (MSC).
fluctuations [39]. Machine learning algorithms analyze and
interpret the spectra, recognizing patterns and identifying char-
acteristics related to food quality or contaminants. Real-time learning model is trained on accurate data.
monitoring allows continuous observation. The IoT aspect con- • Preprocess the cleaned data by normalizing, scaling, or
nects the methane sensor and microcontroller, enabling data transforming it. This step is important to prepare the data
collection and transmission. The system enhances accuracy for the machine learning model.
and reliability, facilitating subsequent analyses or modeling • Train the machine learning model on the preprocessed
processes, and ensuring a smarter approach to food monitoring data. This step involves selecting the appropriate machine
[40]. learning algorithm, tuning its hyperparameters, and eval-
uating its performance.
• Evaluate the performance of the machine learning model
by testing it on new data. This step involves comparing
the predicted values with the actual values and calculating
performance metrics such as accuracy, precision, and
recall.

V. C ONCLUSION
In conclusion, a smart food monitoring system can greatly
improve food safety and reduce food waste by using sensors,
microcontrollers, and wireless communication technologies to
detect and track the quality and freshness of perishable foods.
Such a system can help consumers and food industry profes-
sionals make informed decisions about when to consume or
discard food products, reducing the risk of foodborne illnesses
and minimizing the economic and environmental costs of food
waste. With the advancements in Internet of Things (IoT)
technologies, smart food monitoring systems are becoming
Fig. 2. original spectra (log 1/R) increasingly accessible and affordable, making it possible for
individuals and businesses of all sizes to implement them in
• Collect sensor data for a specific period. This data should their operations. These systems can be customized to suit
include information about food usage, temperature, hu- different needs and requirements and can integrate with other
midity, and other factors that can impact food quality. smart devices and services to create a more comprehensive
• Once the experimental setup is complete, we can start and efficient food monitoring and management system.
the experimental analysis by collecting sensor data and
analyzing it using the machine learning model. The R EFERENCES
following steps can be followed to analyze the sensor
data. [1] A. Rahman, M. J. Islam, A. Montieri, M. K. Nasir, M. M. Reza, S. S.
Band, A. Pescape, M. Hasan, M. Sookhak, and A. Mosavi, “Smartblock-
• Clean the collected data by removing any outliers or sdn: An optimized blockchain-sdn framework for resource management
errors. This step is important to ensure that the machine in iot,” IEEE Access, vol. 9, pp. 28 361–28 376, 2021.
 [2] M. J. Islam, A. Rahman, S. Kabir, M. R. Karim, U. K. Acharjee, M. K. monitoring system in the iot,” IEEE Transactions on Consumer Elec-
Nasir, S. S. Band, M. Sookhak, and S. Wu, “Blockchain-sdn-based tronics, vol. 64, no. 3, pp. 390–398, 2018.
energy-aware and distributed secure architecture for iot in smart cities,” [22] K. Demestichas, N. Peppes, and T. Alexakis, “Survey on security threats
IEEE Internet of Things Journal, vol. 9, no. 5, pp. 3850–3864, 2022. in agricultural iot and smart farming,” Sensors, vol. 20, no. 22, p. 6458,
[3] V. S. Yadav, A. Singh, R. D. Raut, S. K. Mangla, S. Luthra, and 2020.
A. Kumar, “Exploring the application of industry 4.0 technologies in [23] S. Bhushan, B. Bohara, P. Kumar, and V. Sharma, “A new approach
the agricultural food supply chain: A systematic literature review,” towards iot by using health care-iot and food distribution iot,” in 2016
Computers & Industrial Engineering, p. 108304, 2022. 2nd International conference on advances in computing, communication,
[4] M. Faisal, H. Siddiqua, M. J. Islam, and A. Rahman, “An sdn-based se- & automation (ICACCA)(Fall). IEEE, 2016, pp. 1–7.
cure model for iot network in smart building,” in 2022 4th International [24] A. Sheth, U. Jaimini, and H. Y. Yip, “How will the internet of
Conference on Sustainable Technologies for Industry 4.0 (STI), 2022, things enable augmented personalized health?” IEEE intelligent systems,
pp. 1–6. vol. 33, no. 1, pp. 89–97, 2018.
[5] A. Faid, M. Sadik, and E. Sabir, “Iot-based low cost architecture for [25] M. Chen, J. Wan, and F. Li, “Machine-to-machine communications:
smart farming,” in 2020 International Wireless Communications and Architectures, standards and applications,” KSII Transactions on Internet
Mobile Computing (IWCMC). IEEE, 2020, pp. 1296–1302. and Information Systems (TIIS), vol. 6, no. 2, pp. 480–497, 2012.
[6] S. Bakalis, V. P. Valdramidis, D. Argyropoulos, L. Ahrne, J. Chen, P. J. [26] A. Mijanur Rahman, D. H. Kim, H. D. Jang, J. H. Yang, and S. J.
Cullen, E. Cummins, A. K. Datta, C. Emmanouilidis, T. Foster et al., Lee, “Preliminary study on biosensor-type time-temperature integrator
“Perspectives from co+ re: How covid-19 changed our food systems and for intelligent food packaging,” Sensors, vol. 18, no. 6, p. 1949, 2018.
food security paradigms,” Current Research in Food Science, vol. 3, p. [27] M. Bhatia and T. A. Ahanger, “Intelligent decision-making in smart food
166, 2020. industry: quality perspective,” Pervasive and Mobile Computing, vol. 72,
[7] M. A. H. Wadud, T. M. Amir-Ul-Haque Bhuiyan, M. A. Uddin, and p. 101304, 2021.
M. M. Rahman, “A patient centric agent assisted private blockchain on [28] A. Rahman, J. Islam, D. Kundu, R. Karim, Z. Rahman, S. S. Band,
hyperledger fabric for managing remote patient monitoring,” in 2020 M. Sookhak, P. Tiwari, and N. Kumar, “Impacts of blockchain in
11th International Conference on Electrical and Computer Engineering software-defined internet of things ecosystem with network function
(ICECE), 2020, pp. 194–197. virtualization for smart applications: Present perspectives and future
[8] H. Luo, M. Zhu, S. Ye, H. Hou, Y. Chen, and L. Bulysheva, “An directions,” International Journal of Communication Systems, p. e5429,
intelligent tracking system based on internet of things for the cold chain,” 2023.
Internet Research, 2016. [29] B. Yu, P. Zhan, M. Lei, F. Zhou, and P. Wang, “Food quality monitoring
[9] A. Rahman, U. Sara, D. Kundu, S. Islam, M. J. Islam, M. Hasan, system based on smart contracts and evaluation models,” IEEE Access,
Z. Rahman, and M. K. Nasir, “Distb-sdoindustry: Enhancing security in vol. 8, pp. 12 479–12 490, 2020.
industry 4.0 services based on distributed blockchain through software [30] K. M. Shayshab Azad, N. Hossain, M. J. Islam, A. Rahman, and
defined networking-iot enabled architecture,” International Journal of S. Kabir, “Preventive determination and avoidance of ddos attack with
Advanced Computer Science and Applications, vol. 11, no. 9, 2020. sdn over the iot networks,” in 2021 International Conference on Au-
[10] H. Desai, D. Guruvayurappan, M. Merchant, S. Somaiya, and tomation, Control and Mechatronics for Industry 4.0 (ACMI), 2021, pp.
H. Mundra, “Iot based grocery monitoring system,” in 2017 fourteenth 1–6.
international conference on wireless and optical communications net- [31] C. Jamroen, N. Yonsiri, T. Odthon, N. Wisitthiwong, and S. Janreung,
works (WOCN). IEEE, 2017, pp. 1–4. “A standalone photovoltaic/battery energy-powered water quality mon-
[11] H. Kaur, “Modelling internet of things driven sustainable food security itoring system based on narrowband internet of things for aquaculture:
system,” Benchmarking: An International Journal, vol. 28, no. 5, pp. Design and implementation,” Smart Agricultural Technology, vol. 3, p.
1740–1760, 2021. 100072, 2023.
[12] T. Baranwal, P. K. Pateriya et al., “Development of iot based smart se- [32] A. Rahman, C. Chakraborty, A. Anwar, M. Karim, M. Islam, D. Kundu,
curity and monitoring devices for agriculture,” in 2016 6th International Z. Rahman, S. S. Band et al., “Sdn–iot empowered intelligent frame-
Conference-Cloud System and Big Data Engineering (Confluence). work for industry 4.0 applications during covid-19 pandemic,” Cluster
IEEE, 2016, pp. 597–602. Computing, vol. 25, no. 4, pp. 2351–2368, 2022.
[33] A. Rahman, M. J. Islam, S. S. Band, G. Muhammad, K. Hasan,
[13] A. Rahman, A. Montieri, D. Kundu, M. Karim, M. Islam, S. Umme,
and P. Tiwari, “Towards a blockchain-sdn-based secure architecture
A. Nascita, A. Pescapé et al., “On the integration of blockchain and sdn:
for cloud computing in smart industrial iot,” Digital Communications
Overview, applications, and future perspectives,” Journal of Network and
and Networks, vol. 9, no. 2, pp. 411–421, 2023. [Online]. Available:
Systems Management, vol. 30, no. 4, pp. 1–44, 2022.
https://www.sciencedirect.com/science/article/pii/S2352864822002449
[14] M. A. A. Noman, M. J. Hossain, M. M. Kalimulla, S. Azad, and [34] A. Rahman, K. Hasan, D. Kundu, M. J. Islam, T. Debnath, S. S.
M. A. H. Wadud, “An intelligent application for preventing the coun- Band, and N. Kumar, “On the icn-iot with federated learning integration
terfeit medicines through a distributed blockchain,” in 2021 3rd Inter- of communication: Concepts, security-privacy issues, applications, and
national Conference on Sustainable Technologies for Industry 4.0 (STI), future perspectives,” Future Generation Computer Systems, vol. 138, pp.
2021, pp. 1–6. 61–88, 2023.
[15] M. A. Hussen Wadud, A. Rahman, M. J. Islam, T. M. Amir-Ul- [35] M. J. Hossain, M. A. H. Wadud, A. Rahman, J. Ferdous, M. S. Alam,
Haque Bhuiyan, M. J. Hossain, and R. Hossen, “A decentralized T. M. Amir Ul Haque Bhuiyan, and M. F. Mridha, “A secured patient’s
secure blockchain-based privacy-preserving healthcare clouds and ap- online data monitoring through blockchain: An intelligent way to store
plications,” in 2023 International Conference on Electrical, Computer lifetime medical records,” in 2021 International Conference on Science
and Communication Engineering (ECCE), 2023, pp. 1–6. Contemporary Technologies (ICSCT), 2021, pp. 1–6.
[16] M. A. H. Wadud and M. R. H. Rakib, “Text coherence analysis based [36] M. J. Hossain, M. A. H. Wadud, and M. Alamin, “Hdm-chain: A secure
on misspelling oblivious word embeddings and deep neural network,” blockchain-based healthcare data management framework to ensure
International Journal of Advanced Computer Science and Applications, privacy and security in the health unit,” in 2021 5th International
vol. 12, no. 1, 2021. Conference on Electrical Engineering and Information Communication
[17] S. I. Khan, A. Shahrior, R. Karim, M. Hasan, and A. Rahman, “Multinet: Technology (ICEEICT), 2021, pp. 1–6.
A deep neural network approach for detecting breast cancer through [37] S. U. Shariff, M. Gurubasavanna, and C. Byrareddy, “Iot-based smart
multi-scale feature fusion,” Journal of King Saud University-Computer food storage monitoring and safety system,” in International Conference
and Information Sciences, vol. 34, no. 8, pp. 6217–6228, 2022. on Computer Networks and Communication Technologies: ICCNCT
[18] M. A. H. Wadud, M. Mridha, and M. M. Rahman, “Word embedding 2018. Springer, 2019, pp. 623–638.
methods for word representation in deep learning for natural language [38] M. S. I. Khan, A. Rahman, T. Debnath, M. R. Karim, M. K. Nasir,
processing,” Iraqi Journal of Science, pp. 1349–1361, 2022. S. S. Band, A. Mosavi, and I. Dehzangi, “Accurate brain tumor detection
[19] W. K. Mok, Y. X. Tan, and W. N. Chen, “Technology innovations for using deep convolutional neural network,” Computational and Structural
food security in singapore: A case study of future food systems for an Biotechnology Journal, vol. 20, pp. 4733–4745, 2022.
increasingly natural resource-scarce world,” Trends in food science & [39] X. Li, T. Wang, X. Lu, W. Li, N. Zhu, M. Song, C. Xu, Q. Yu, and
technology, vol. 102, pp. 155–168, 2020. X. Yu, “Integration of spectroscopy, iot, and machine learning for food
[20] H. Alkahtani and T. H. Aldhyani, “Developing cybersecurity systems quality monitoring,” Journal of Food Engineering, vol. 285, p. 110088,
based on machine learning and deep learning algorithms for protecting 2020.
food security systems: industrial control systems,” Electronics, vol. 11, [40] J. Luo, Q. Zhang, L. Wang, P. Li, Y. Liu, S. Chen, and Z. Li, “Smart food
no. 11, p. 1717, 2022. monitoring system integrating spectroscopy, iot, and machine learning
[21] P. Sundaravadivel, K. Kesavan, L. Kesavan, S. P. Mohanty, and for quality assessment,” Food Control, vol. 130, p. 108225, 2021.
E. Kougianos, “Smart-log: A deep-learning based automated nutrition

View publication stats